A paper for the New Zealand Planning Council conference, “The Distribution of Income and Wealth in New Zealand”, by Brian Easton*
Keywords: Distributional Economics; Social Policy;
Probably the first text I ever read as an economics undergraduate was Tjalling Koopmans’ “Three Essays on the State of Economic Science” (1957). The book contains an elegant account of general equilibrium economics and how the price mechanism leads to an efficient competitive equilibrium. There are some complications, one of which Koopmans deals with succinctly.
“One ‘hard boiled’ alternative would be to assume instantaneous elimination by starvation of those whose resources prove insufficient for survival…. An alternative … would be to recognise the existence of income transfers through taxation and social insurance.” (p. 62)
From the beginning I have never doubted that the market is incapable of ensuring a living wage. As Koopmans says, there will be people with insufficient physical, financial and human resources who will starve if dependent upon the market.
Thus, the market is unable to deliver a living wage to all. However, we may want to ask to what extent the market can provide a living wage. This involves both a matter of definition, and a view of the empirical evidence. It also includes some review of the market-related approach mentioned in the Koopmans quotation, sometimes called the “social wage”.[1] In order to be comprehensive in the time at my disposal, I shall have to be brief on each topic.[2]
The Living Wage
We shall take the concept of a living wage to mean an adequate standard of living. In doing so, the focus is shifted in two ways. First, the concern is with income from all sources, and not just from labour effort. Second, the adequacy is not merely to prevent starvation and maintain life at some minimalist, perhaps animal, level but to focus upon living in a community.
The best definition New Zealanders have of this adequate standard of living comes from the Report of the Royal Commission on Social Security (1972) who wrote that the community should aim
“First, to enable everyone to sustain life and health;
“Second, to ensure, within limitations which may be imposed by physical or other disabilities, that everyone is able to enjoy a standard of living much like that of the rest of the community, and thus is able to feel a sense of participation in and belonging to the community;
“Third, where income maintenance alone is sufficient (for example, for a physically disabled person), to improve by other means, and as far as possible, the quality of life available.” (p. 65)
The point of this list is that the Royal Commission did not have in mind just the maintenance of body and soul, but as indicated in the second objective, an aim to enable everyone to enjoy a sense of participation in and belonging to the community.
Note that the Royal Commission’s notion is closely related to that found in John Rawls’ “Theory of Justice” (1971) of the fundamental primary good being “self-respect”, which in turn is related to the traditional New Zealand welfare concern of “dignity”. One is the objective from a community point of view, the other from an individualistic point of view. A person who has self-respect will be able to participate in and feel they belong to the community.
There are a number of crucial consequences to this approach. The first is that the minimum standard of living is not fixed eternally, but changes through time and between communities. It is not a notion of absolutes.
Second, the notion is not merely an economic one, referring only to the material standard of living. There are other elements which are outside the compass of an economist – perhaps to be dealt with by law, education, attitudinal change, and so on. I give but one example: about a decade ago women told men that the practice of asking masculine nouns and pronouns to also encompass women made them feel as if not belonging and unable to participate fully. As a result there has been an adaptation of the language to recognise explicitly both sexes, and an associated, but somewhat slower, change in attitudes.
This does not mean that the economist can ignore all the social issues where wage determination is involved. For instance, equal pay for women is a part of their self-respect. It may be much easier and cheaper to legislate for equal pay, than to use some more complex approach which increases wage flexibility, but involves some other means of maintaining women’s self-respect via, perhaps, a financial grant.
Another area where economists need to pay attention is in that of market wealth distribution, which has social and political implications different from the income distribution. To continue the theme of the above illustrations, it is a fact that women still have a less than equal share of the nation’s private wealth.
The third point is that there is almost certainly no unique scientifically determined level for the acceptable minimum material standard of living. The level which is chosen will involve a series of judgements, which can be debated vigorously. This is not to conclude that no figure can be settled upon, and, if surveys of attitudes of New Zealanders are any guide, the likely range for the debate is probably quite narrow.
In practice the acceptable minimum material standard of living used in New Zealand has been that set by the Royal Commission on Social Security for a married couple on the standard social security benefit. Subsequent work suggested that it was about right, although we could easily argue 10 percent either way. Note that the level for a married couple, often called the Benefit Datum Line (or Poverty Line) has to be translated to an equivalent level for households other than a couple using household equivalence scales. Even more problematic is that there has been little systematic work since the mid-1970s, and subsequent changes may mean that the standard is out of line, in either direction. But until there is this systematic work, we are stuck with the assessment of the 1972 Royal Commission.
For Whom the Market Fails
Given this measure of the material standard of living, we ask for whom does the market fail to provide. An answer to this question can be derived from the work of Suzanne Snively based upon 1981/2 data (1987). The overall conclusion is 30.3 percent of the population had a market income below the benefit datum level.[3]
We have to be careful interpreting this figure because the market incomes occurred in an economy in which there was considerable welfare provision. If that provision had not been there, people would have adjusted their behaviour. Those with low market income supported by social security benefits might have found work or have saved more for retirement; some of those living would have died if there had been no welfare state, and so on. The figure is an indication of the proportion of the community who, if the welfare state were abolished overnight, would wake up in the morning with market incomes that would not sustain a minimum standard of living. Thus the figure of a quarter for whom the market is not providing should be treated as an order-of-magnitude figure, with a very limited meaning.
There are four broad reasons for being in this group; handicap, retirement, unemployment and youth. Handicap may arise for physical, intellectual, or psychological reasons. The handicap may exist at birth or arise as a consequence of a subsequent event. In some cases, a market mechanism, such as insurance, can provide for some of the handicapped, but in general it cannot. Unless there is some social provision, the handicapped are obvious candidates for the starvation to death category.
The issue of the retired is more complicated. Since at least 1898 New Zealand has had a state-provided unfunded welfare provision for the elderly, which has become increasingly comprehensive and generous. If no such scheme had been instituted, or a different approach had been taken, it is likely that today there would be greater market provision for the retired. For instance in many countries there are state-administered, compulsory schemes, funded by a levy during the retiree’s working life, but which nevertheless receive injections of taxpayer funds. In international comparisons, their pensions are treated as market provision, which is one of the reasons why the distribution of market income in New Zealand can appear more unequal.
It does not follow that voluntary private provision for retirement will guarantee an adequate income. Indeed, the three great transformations in public provision for retirement occurred following a major failure in the private system (Easton 1981). It is no accident that the 1898 Old Age Pension Act and the 1938 Social Security Act followed periods of long depression, which have the effect of wiping out private savings. Similarly, the upheaval in the early 1970s followed the shift into double digit annual inflation, which also wiped out much of the value of many people’s savings. Even if there were no problems of transition, it would be foolish to assume that a switch to a private market scheme would result in a retirement provision as comprehensive and adequate as schemes with public involvement.
In an obvious sense, the market has failed the unemployed – in two ways. First they do not receive market income – nowhere in the world is there an entirely private-based insurance scheme to supplement the incomes of the unemployed. Second, unemployment means diminished self-respect and less of a feeling of participation and belonging.
However, the argument that the market has failed to provide jobs for the unemployed needs to be probed. It is argued that various unjustified government interventions mean that the unemployed cannot get jobs. A frequent claim is that minimum wage laws prevent them from obtaining jobs by offering lower wages. The argument goes on that if there was a fall in the relative wage rates for those with skills that are not currently strong demand, there would be an increase in the total number of jobs. It acknowledged that there will be some job displacement – that is, others will lose thei jobs as a result of these changes – but in aggregate there will be an increase in th• total number of jobs and a reduction in the total unemployed.
Even if this were so, there would remain the issue as to whether the income of the unemployed would be higher. To make my argument transparent, I shall assume there is an unemployment benefit. Nevertheless, I assert that under some circumstances workers receive higher market incomes from a higher wage rate, which results in som, unemployment. What happens is that workers can receive more income from periods unemployment and higher wages while they are working, than from full employment with lower wages. Moreover, I can report to you that the labour market conditions for this theoretical possibility actually exist in Australia and New Zealand (Easton 1987).
This does not mean that there is no case for increased labour market flexibility: Rather, it means that increases in relative wage flexibility will not deliver all that their advocates seem to promise. Moreover, if the advocates were to use a more sophisticated analysis they would notice that while many workers would be worse off income terms as a result of increased relative wage flexibility, the rest of us • investors and those on high wages and salaries – would be better off. This suggest! policy conclusions which I have detailed elsewhere, that show how to attain higher employment and higher incomes for everyone by coupling increased wage flexibility with lower taxation on workers and higher rates on the rich (Easton 1987).
This is, of course, not a purely market solution. It is almost certain that the marker by itself will be unable to deliver a decent living wage to those prone to beirql unemployed.
The last category which tends to have inadequate market incomes is the young, or till young and their parents. In the spirit of many of those who advocate labour market liberalization, it could be argued that poverty arising from bringing up children is consequence of unnecessary government intervention. There are laws which limit child employment and require compulsory attendance at school, thus preventing familiesnrecovering the costs of child-rearing by sending them out to work.
An even more onerous intervention is the prohibition of slavery. This prevents parents from contracting with finance companies to fund the child-rearing in return for the corporation receiving a share of the child’s earnings when they become a working adult.
This illustrates two important principles. First, private markets can solve certain social problems, but the solution may be shocking and unacceptable – like enslavement or death from starvation. Second, many of the problems associated with the economics of children are investment ones, which are not readily tractable to the conventional approach of investment in physical assets.
Family Policy
The issue of how to deal with the income problem of families with children has a long history in New Zealand, as elsewhere. Pember Reeves seems to have ignored the issue but following the “Harvester” decision of Justice Henry Higgins in the Australiar Arbitration Court in 1907, the New Zealand Court of Arbitration set a minimum wag( based on the perceived needs of family. Higgins was explicit, spelling out the right: of the worker as “a human being in a civilized community entitled to marry and raise family”, and he settled for a minimum wage as being sufficient for a family of five. (Holt 1987).
By 1922, Judge Frazer of the New Zealand Court of Arbitration was pondering about the feasibility of the strategy. If all workers are supporting five people it may be successful but if households consist of one to twelve members, have one to three earners, have children of different ages with different maintenance costs, and have markedly different housing costs, it is not obvious that a single wage can meet all the needs. A wage adequate for a family of five will be luxurious for a single person, and penurious for a family of ten.
Step by step, government assistance for families was introduced: in 1914 tax exemptions for children, in 1927 the first incomes-tested family allowance. The 1938 Social Security Act made no direct income provision for children in working families. It appears that it was thought that full employment, together with an appropriate minimum wage, free education and medical services, and cheap housing, was considered sufficient to meet family needs; a similar, view, I suspect, to that of Pember Reeves.
However, in 1946 the universal family benefit was introduced, and in the late 1940s there was further minor financial relief to families through free services. This approach which was carried on during the three-year term of the second Labour government, after which the thrust for a conscious family policy seems to have been lost.
The 1972 Royal Commission on Social’ Security did not really address family policy issues. At about this time the conventional wisdom held that the poverty problem was limited to beneficiaries. However, from 1976 the third National Government began experimenting with income tax packages to meet family needs, an approach continued by the current Labour Government.
1979 was perhaps the last opportunity the Court of Arbitration had to consider the notion of the minimum living wage, its authority to do so being removed by the Remuneration Act in that year. It is fascinating to speculate how the Court may have approached the issue, perhaps there would have been two covert and one overt argument for the determination.
The covert ones need to be recounted, but briefly. First, union militancy in wage determination is often based upon those workers most financially squeezed leading the shop floor charge. If they are respected married men with children, such pressures are reinforced. Family income maintenance packages which alleviate these pressures also alleviate such militancy. Second, the notion that the father is responsible for maintaining the family through his own labour is still a deep part of the psyche of married men and women. It is sometimes called a “fertility and finance” attitude, or if you want to trivialize it, “semen and cents”, and contrasts with the view that fathers are also involved with emotional, educational, and recreational activities of the family. The attitude is dying, but slowly, and remains important in parts of the population.
The overt issue the Court of Arbitration would have had to confront was the view that it was the responsibility of the employer to pay a decent wage, and family maintenance packages based upon taxation and benefits were a means for employers to avoid this responsibility.
The Court would have had some difficulty denying this case in strictly judicial terms. The principle already existed as precedent in earlier cases of the Court. Moreover, for a number of other reasons the Court had been diffident in the past about taking taxation into consideration when setting wages. How the Court would have finally decided can only be conjectured – one would expect that common sense would have interpreted the law judicially – but we can give the account of a mainline economist. The first point would be to ask what is the primary function of wages. The approach of James Meade is to focus on their role as providing employment, and their objective to maintain as full employment as possible (Meade, 1982, Philpott 1985). It thereby denies that the primary purpose of wages is to generate individual household income. That this occurs is felicitous; that it occurs rather unsatisfactorily is not surprising.
Separating the role of wages-as-costs, and hence as a major determinant of employment, from the role of wages-as-income, and hence as a major determinant of the material standard of living, is possible because of the existence of a tax and benefit redistributive mechanism. It is also necessary, given the more complex social objectives and conditions which exist today.
The economist then focuses upon the feasibility of using the wage system to provide adequate incomes, noting that the dramatic social changes which have occurred over the last twenty years would frustrate any attempt to pursue this goal. The most obvious is the rising importance of mothers in the work force, but other factors include increasing part-time work, higher unemployment, changing family sizes (including more couples choosing not to have children), solo parent families, and the unsatisfactory state of housing finance. Such social changes provided the pressure which resurrected the interest in family policy in the 1970s.
The third general point an economist would make is that tax concessions, social security benefits and publically-provided services are not subsidies to employers. It appears that the majority of welfare distribution in New Zealand is not vertical but horizontal (including inter-temporal horizontal which will not be developed further here). Horizontal redistribution occurs when there are transfers within a group. As it happens, workers are both the main source of tax revenue, and the main recipients of the welfare state. The sick worker whose treatment is paid for by a state medical benefit, the worker with children in state education, those on the unemployment benefit and so on, are having their welfare primarily funded from taxation paid by other workers. There is some vertical redistribution, from the rich to the poor, but this horizontal redistribution is more important. Family income maintenance is a part of this redistribution between workers. Thus it is not a subsidy to employers; in a more caring and socially sensitive society it would be called “mutual aid”.
The Social Wage
The technical term for this system of mutual aid is the “social wage” which is usually defined as that part of government spending which provides benefits in cash or kind, to individuals and families. It encompasses government spending on education, health, social security, social welfare, housing, tax rebates and credits for social purposes, and expenditure on community amenities.
Alas, New Zealand has no comprehensive review of its social wage similar to that commissioned by the Australian Economic Planning Advisory Council (1987). However we do know from Suzanne Snively’s research that the New Zealand social wage is significant in cost and in redistributive impact.
Recall that if households had to depend upon only their market income, 30.3 percent of people would have been below the Benefit Datum Level, or Standard Poverty Line. The actual situation was households paying tax, but having their income supplemented by the social wage including largely free medical and educational services, social security benefits, and tax credits. Now only 9.6 percent of people were below the benefit datum level in 1981/2. This is a figure comparable to that calculated by Rochford and Pudney for 1980/1 (1984), if they had included employment costs in their Household Equivalence Scales.
This would give an aggregate number of poor in 1981/2 between 320,000 and 400,000. This figure is lower than the figure I suggested applied for 1974/5 of 550,000 New Zealanders below the B.D.L. (Easton 1976). There are three groups of reasons why there is this difference.
First, the sources of data: the 1974/5 Household Survey was the first of its kind, and the Department of Statistics has improved its techniques since, including a more accurate measurement of income. In addition, there have been some improvements in the measurement technology, including Snively’s ability to anchor back to the 1981 census and the use of the ASSET model to give greater precision to measuring the effects of taxes and transfers.
A second group of reasons centres around government responses to the concern which developed in the mid-1970s about the existence of family poverty. From 1976 the tax system has been evolving to give greater support to poor families, and insofar as this has been successful it should have reduced the number of poor.
Third, social and economic changes may be alleviating poverty. It is true that real after-tax wages have not been rising, while unemployment has. But more mothers are working and families are getting smaller; both phenomena would reduce the amount of poverty.
All these points were made in a paper I gave in 1980. The merit of this recent research is that some of the theoretical issues I raised then have been applied, with the conclusion that the 1974/5 figure is probably too high.
One vital issue I raised in both my 1975 and 1980 papers, and which has not been tackled yet, is housing costs. We still assume that each household has average housing costs. However, housing costs vary enormously and erratically from household to household, and it is not obvious that the number of people below the benefit datum line with low housing costs balance those just above with high housing costs. My 1974/5 estimate included some allowance for more people being depressed by high housing costs than raised by low ones. I look forward to research that will investigate this issue. Another one which deserves more attention is the costs of children of different ages. Most of the research uses Household Equivalence scales which treat all children equally, even though an adolescent is much more expensive to keep than an infant. So I do not think the new evidence may be interpreted to suggest there has been a dramatic change in the level of poverty between 1984/5 and 1981/2. As I have always insisted, the figures give orders of magnitude only.
It is also important that the 1981/2 figure not be used as a proxy for the current level of poverty in New Zealand. Since 1981/2 there have been falling after-tax wages particularly it would appear for the poorest; sharply rising housing costs for some; and rising unemployment. Whether this unemployment is significant among the main earners of households is not known.
Note too that, despite family support, a two-child family with a single earner on the average wage gets less relative support than when the family benefit was introduced in 1946.
The above analysis suggests a much lower figure for poverty in New Zealand than the estimate proposed by Waldegrave and Coventry (1987). I can identify three major differences.
First, they argue that the true poverty line is above the BDL.
Second they argue that poverty has been rising rapidly in recent years, paralleling the rise in unemployment.
Third, they use a wider definition of poverty. The studies quoted above measure only those people in households with incomes below the poverty line. It would be in the spirit of the Royal Commission on Social Security that the poor also include those who do not receive a fair share of the household income, those who are unemployed but in higher income households, those who are sick but cannot get access to medical services, those who belong to cultural minorities (and we might include women here) and feel oppressed, those whose housing or wealth holdings are inadequate, and so on.
These factors would all validly increase the number of the poor, although I am inclined to the view that the Waldegrave/Coventry estimate is probably a little high.
Poverty in New Zealand
The material on poverty provides some important conclusions. First, despite the social wage there are hundreds of thousands of New Zealanders who are poor in that they postpone visits to the doctor, they cannot afford to keep their house warm in winter, they have few if any holidays away from home, and they occasionally miss meals or have meatless meals because of food costs.
Second, the Snively work indicates nevertheless that the social wage was pretty effective at reducing poverty in New Zealand. Without it, 30.3 percent of the population would be poor if they were to continue in their present circumstances. With it, that figure reduces to 9.6 percent. Any attempt to rejig the wage structure to provide a living wage would not be so successful.
Third, and here I may be asking the study to bear more weight than was intended, the Snively work does suggest that many of the poor faced high effective tax rates. At that time, a top income recipient faced a maximum tax rate of 60 percent; many poor faced even higher rates, in terms of the degree to which their social wage reduced as their market income rose.
This suggests that the effectiveness of the social wage could be improved, but to do so would require a more detailed analysis than is currently available.
Policy Conclusion
This paper may be summarised by a set of policy conclusions.
First, given the sort of political and social structure that we have, and the sort of economic options that it generates, it is most unlikely that the market system by itself will guarantee everyone a decent living wage. It needs some modification.
Second, the primary objective of the wage structure should be to maximise employment with a high standard of living. This objective is a priority because
(i) employment gives self-respect, and a sense of participating in and belonging to a community;
(ii) full employment gives greater national output and the possibility of a higher social wage;
(iii) for most, but not all, the income from employment is a substantial (and often adequate) means of attaining the social wage.
Third, while maximum employment is the priority for wage determination, other objectives such as status enhancement and income maintenance should be included providing they are not too damaging to the primary objective.
Fourth, while excessive wage rigidity is to be eschewed, relative wage flexibility is not likely to solve the employment problem, and it is possible that excessive pursuit of wage flexibility is likely to damage the status enhancement and incomes maintenance objectives without markedly adding to jobs.
Fifth, an integral step in abandoning the objective of guaranteeing a living wage through the employment system is a fair social wage which results in an overall income maintenance system which guarantees an adequate minimum standard of living.
Sixth, on the basis of available information, the existing social wage, which consists most obviously of social security and income tax credits plus reasonably free education, health and other welfare services, is fairly successful at reducing much of the poverty which would exist in New Zealand if there were no social wage.
Seventh, there is, however, no room for complacency. There are hundreds of thousands of people who have a standard of living which means they cannot feel they belong to the community, nor participate in it.
Eighth, there is, moreover, probably room for improvements in the efficiency of the delivery of the social wage. There are three obvious areas for attention. One is the high effective rates of taxation on the poor – I suggest a good policy principle is that top tax rates on the poor and on the middle income range should not exceed the top tax rates on the rich. The second area for urgent attention is housing. And third is the infringement upon the traditional scope of the welfare state by higher charges to users and victims which appear to be aimed at reducing funding costs rather than improving the efficiency of the social wage.
Ninth and finally, I do not think that a major reconstruction of the social wage is urgent. A more sensible approach would be to commence with a well founded independent research program, to measure and assess the effectiveness of the social wage. Without such a research program, any reconstruction may add to poverty and inefficiency; and in any case there will be widespread additional mistrust of the reformers.
Notes:
1.David Pichaud reminded us during the seminar that there are also considerable
voluntary transfers between individuals, most notably within the family.
2. The arguments are elaborated in the author’s Social Policy and the Welfare
State in New Zealand (Allen and Unwin 1980), and Wages and the Poor (Allen and Unwin, 1986).
3. I am grateful to Suzanne Snively for access to her research (1987). A paper detailing the method and elaborating the results will be made available as soon as the Trust has the resources.
Bibliography
Easton, B. H. (1976) “Poverty in New Zealand; Estimates and Reflections” Political Science, December 1976. (Reprinted in B. H. Easton, Income Distribution in New Zealand, NZIER Research Paper 28, 1983)
(1980) Poverty in New Zealand; Five Years After (Paper to conference of NZ Sociological Association, 1980 – Mimeo)
(1981) Pragmatism and Progress: Social Security in the Seventies (CUP)
(1987) Labour Flexibility, Wages and Free Lunches (University of Melbourne Department of Economics, Research Paper No. 180)
Economic Planning Advisory Council (1987) Aspects of the Social Wage: A Review of Social Expenditures and Redistribution (Australian Government Printers)
Holt, J. (1987) Compulsory Arbitration in New Zealand: The First Forty Years (AUP).
Koopmans, T. J. (1957) Three Essays on the State of Economic Science (McGraw Hill).
Meade, J. E. (1982) Stagflation (Vols I & II) (Allen and Unwin)
Philpott, B.P. (1985) Economic Research and Economic Policy (NZIER Discussion Paper 29)
Rawls, J. (1971) A Theory of Justice (Harvard University Press).
Rochford, M. W. and K. J. Pudney (1984) An Exploratory Application of Income Equivalences to the Examination of Household Living levels (Working Paper 006R/17P of the Department of Social Welfare)
Royal Commission on Social Security (1972) Social Security in New Zeaand (Government Printer)
Snively, S. (1987) The 1981/82 Government Budget and Household Income Distribution (NZ Planning Council).
Waldegrave , C. and R. Coventry (1987) Poor New Zealand : An Open Letter on Poverty (Platform Publishing).
*This paper is sponsored by the Economic and Social Trust of New Zealand. The primary objective of the Trust is to promote independent quality economic and social research in the public interest. The trust is grateful for support from the Downing Fellowship in Social Economics at the University of Melbourne, the Research Project on Economic Planning of the Victoria University of Wellington, and the Easton family.